paper-with-me

Papers

How Well Can Transformers Emulate In-context Newton's Method?

2024-03-05 · Angeliki Giannou, Liu Yang, Tianhao Wang, Dimitris Papailiopoulos, Jason D. Lee

Transformer-based models have demonstrated remarkable in-context learning capabilities, prompting extensive research into its underlying mechanisms. Recent studies have suggested that Transformers can implement first-order optimization algorithms for in-context learning and even second order ones for the case of linear regression. In this work, we study whether Transformers can perform higher order optimization methods, beyond the case of linear regression. We establish that linear attention Transformers with ReLU layers can approximate second order optimization algorithms for the task of logistic regression and achieve $\epsilon$ error with only a logarithmic to the error more layers. As a by-product we demonstrate the ability of even linear attention-only Transformers in implementing a single step of Newton's iteration for matrix inversion with merely two layers. These results suggest the ability of the Transformer architecture to implement complex algorithms, beyond gradient descent.

📄 PDF Abstract BibTeX arXiv:2403.03183

Code (0)

등록된 구현이 없습니다.

Tasks

In-Context Learningregression

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Multi-Head Attention 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

Transformers Learn to Achieve Second-Order Convergence Rates for In-Context Linear Regression

2023-10-26 · Deqing Fu, Tian-Qi Chen, Robin Jia, Vatsal Sharan

Transformers excel at in-context learning (ICL) -- learning from demonstrations without parameter updates -- but how they do so remains a mystery. Recent work suggests that Transformers may internally run Gradient Descen…

In-Context Learning

Newton Raphson Emulation Network for Highly Efficient Computation of Numerous Implied Volatilities

2022-10-28 · Geon Lee, Tae-Kyoung Kim, Hyun-Gyoon Kim, Jeonggyu Huh

In finance, implied volatility is an important indicator that reflects the market situation immediately. Many practitioners estimate volatility using iteration methods, such as the Newton--Raphson (NR) method. However, i…

Deep Learning

Looped Transformers as Programmable Computers

2023-01-30 · Angeliki Giannou, Shashank Rajput, Jy-yong Sohn, Kangwook Lee 외

We present a framework for using transformer networks as universal computers by programming them with specific weights and placing them in a loop. Our input sequence acts as a punchcard, consisting of instructions and me…

In-Context Learning

AI-Newton: A Concept-Driven Physical Law Discovery System without Prior Physical Knowledge

2025-04-02 · You-Le Fang, Dong-Shan Jian, Xiang Li, Yan-Qing Ma

Current limitations in human scientific discovery necessitate a new research paradigm. While advances in artificial intelligence (AI) offer a highly promising solution, enabling AI to emulate human-like scientific discov…

scientific discovery

A Closer Look at In-Context Learning under Distribution Shifts

2023-05-26 · Kartik Ahuja, David Lopez-Paz

In-context learning, a capability that enables a model to learn from input examples on the fly without necessitating weight updates, is a defining characteristic of large language models. In this work, we follow the sett…

In-Context Learning